get_stats-ConsensusPartition-method.RdGet statistics
# S4 method for ConsensusPartition
get_stats(object, k = object@k, all_stats = FALSE)A ConsensusPartition-class object.
Number of subgroups. The value can be a vector.
Whether to show all statistics that were calculated. Used internally.
The statistics are:
1 - proportion of ambiguous clustering, calculated by PAC.
The mean silhouette score. See https://en.wikipedia.org/wiki/Silhouette_(clustering) .
The mean probability that each partition fits the consensus partition, calculated by concordance.
The increased area under eCDF (the empirical cumulative distribution function) curve to the previous k.
This is the percent of pairs of samples that are both in a same cluster or both are not in a same cluster in the partition of k and k-1. See https://en.wikipedia.org/wiki/Rand_index .
The ratio of pairs of samples are both in a same cluster in the partition of k and k-1 and the pairs of samples are both in a same cluster in the partition k or k-1.
A matrix of partition statistics.
data(golub_cola)
obj = golub_cola["ATC", "skmeans"]
get_stats(obj)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 1.0000000 0.9953240 0.9977778 0.50705126 0.4933490 0.4933490
#> 3 3 1.0000000 0.9788419 0.9900000 0.26217651 0.8595462 0.7182104
#> 4 4 0.8024583 0.8560353 0.9047222 0.11098757 0.9104069 0.7553419
#> 5 5 0.7633889 0.6659601 0.8386111 0.05163403 0.9624413 0.8714859
#> 6 6 0.7295874 0.5711678 0.8069444 0.03855567 0.9139280 0.7002725
get_stats(obj, k = 2)
#> k 1-PAC mean_silhouette concordance area_increased Rand Jaccard
#> 2 2 1 0.995324 0.9977778 0.5070513 0.493349 0.493349